Efficiency enhancement of energy supply chains using a machine learning-driven network evaluation framework for blockchain adoption

The Oil and Gas (O&G) sector is essential to the global energy supply, yet it still struggles with inefficiencies caused by conventional, paper-driven decision-making processes in its supply chain. This study aims to assess the integration of blockchain technology as a strategic solution to these challenges. It examines strategies for blockchain integration considering organizational, supply chain, technological, and external factors. A novel approach based on four network-based data-driven optimization models is utilized to investigate a case study in Norway's O&G sector over a five-year period (April to October 2020). These models incorporate uncertainty to assess the effectiveness of various blockchain integration strategies. The findings demonstrate that "single-use strategy" is effective in low-risk scenarios, while "substitution strategy" is optimal in high-risk scenarios. The application of the Super-Efficiency Network Data Envelopment Analysis (SENDEA) model represents its robustness in evaluating blockchain adoption within the industry. Furthermore, a an efficient Machine Learning (ML) technique; i.e., Principal Component Analysis (PCA), is utilized to further assess blockchain adoption strategies. Therefore, policymakers are recommended to support risk-based blockchain adoption by directing organizations to implement single-use strategies in low-risk environments and substitution strategies in high-risk circumstances. Additionally, they can take advantage of practical evaluation frameworks provided by the SENDEA and uncertain models to systematically appraise and prioritize blockchain strategies based on their potential impacts.